Factorized-VITS: Decoupling Prosody and Text in End-to-End Speech Synthesis without External or Secondary Aligner.

Yining Liu, Alexander Waibel

Browse the full ICASSP paper archive.